I want to use Ansys fluent to calculate the temperature field and velocity field. My model contains a sealed box in which is 1/3 medium and 2/3 air. I set a constant energy source in medium（ regard as heating the medium） then I obtain the result of temperature field and velocity field . But I have trouble with the velocity field result. In theory, medium and air shall have the same speed at their interface. But in fluent, when I set a wall between medium and air, no slip (shear condition) resulted in both zero speed, specified shear(shear condition) with zero shear stress(free slip) resulted in diffident speed. When I do not set the wall, the result is just a disaster. So I wonder where can I set the air-medium boundary to ensure they have the same speed at interface after heating. Now I use k-epsilon/ laminar model to calculate. I have tried multiphase VOF but failed.
can you show pictures of your model? Also what do you mean by the term 'medium' ?
Thanks. The model is simple. There are only three parts, shell, air, medium. Medium is cell culture medium but you can regard it as water. I think it is a little bit similar to model of natural heat convection. Temperature change leads to density variation, then medium and air begin to flow at a quite low speed. My boss insisted that the velocity is constant between air and medium. But I don't know how to realize this in fluent.
you do not have set a wall between the medium and air. Try reducing your timetep size and also make sure you have fine mesh at the boundary where heat is applied.
You may use an experimental method before any numerical simulation like "Laser Doppler Velocimetry" or "Infrared Camera" or PIV. It may help you for setting the gas-liquid boundary.
It depends on which model you are using. There are two main models for simulating two-phase flows: VOF and DPM
Now I think the only method to solve this problem is to use the specific shear and give the correct shear stress.
When I use VOF model, maybe because I have to use boussinesq model for both air and medium.(sealed box, velocity only depends on temperature just like natural heat convection.). At beginning, the medium and air both have 0.5volume fraction. But after heating, the water is nearly disappear. It seems like air compress water to a very very small volume fraction.
Are you sure that considering the volume fraction of 0.5 is correct for them? I think it is wrong. According to the figure you have attached, I think the volume fraction should be 1.
According to the figure, it seems that air and medium enter from a separate entrance (they are not mixed). Therefore, the volume fraction of each is 1.
Thanks. In fact, I mean air and medium both have 1/2 part of the box. I adapt two regions in sealed box first then patch one as air, the other is medium. I think because their thermal expansion rate is different, so after heating because the medium have a smaller thermal expansion rate, it will be suppressed by air to a small volume fraction.
Thank you for your explanation. To be honest, your explanation does not make sense to me. I feel that you have not set the boundary conditions correctly and that is what caused the strange results.
If you have 1 inlet and 2 types of fluids entering from it, then you must determine the volume fraction according to the mass rate of both of them, but the total volume fraction must be 1. For example, 0.5 and 0.5 or 0.3 and 0.7. These ratios depend on their mass flow rate.
But if you have 2 inlets and only one type of fluid enters from each of inlets, then the volume fraction for inlet 1 will be 0 and 1 and for inlet 2 it will be 1 and 0.
In my opinion, this is a static heating problem, so I do not use inlets. Air and medium are zero speed at first. They will have little speed because I am heating them.
Just like heating water in a sealed container, but in my case, temperature changed only from 0℃ to about 1℃， so I needn't care about water vapor.
- Badges
- Science topic